Roy C. Tew and Rodger W. DysonGlenn Research Center, Cleveland, Ohio

Scott D. Wilson and Rikako DemkoSest, Inc., Middleburg Heights, Ohio

Overview 2004 of NASA-Stirling ConvertorCFD Model Development and RegeneratorR&D Efforts

NASA/TM—2004-213404

November 2004

The NASA STI Program Office . . . in Profile

Since its founding, NASA has been dedicated tothe advancement of aeronautics and spacescience. The NASA Scientific and TechnicalInformation (STI) Program Office plays a key partin helping NASA maintain this important role.

The NASA STI Program Office is operated byLangley Research Center, the Lead Center forNASA’s scientific and technical information. TheNASA STI Program Office provides access to theNASA STI Database, the largest collection ofaeronautical and space science STI in the world.The Program Office is also NASA’s institutionalmechanism for disseminating the results of itsresearch and development activities. These resultsare published by NASA in the NASA STI ReportSeries, which includes the following report types:

• TECHNICAL PUBLICATION. Reports ofcompleted research or a major significantphase of research that present the results ofNASA programs and include extensive dataor theoretical analysis. Includes compilationsof significant scientific and technical data andinformation deemed to be of continuingreference value. NASA’s counterpart of peer-reviewed formal professional papers buthas less stringent limitations on manuscriptlength and extent of graphic presentations.

• TECHNICAL MEMORANDUM. Scientificand technical findings that are preliminary orof specialized interest, e.g., quick releasereports, working papers, and bibliographiesthat contain minimal annotation. Does notcontain extensive analysis.

• CONTRACTOR REPORT. Scientific andtechnical findings by NASA-sponsoredcontractors and grantees.

• CONFERENCE PUBLICATION. Collectedpapers from scientific and technicalconferences, symposia, seminars, or othermeetings sponsored or cosponsored byNASA.

• SPECIAL PUBLICATION. Scientific,technical, or historical information fromNASA programs, projects, and missions,often concerned with subjects havingsubstantial public interest.

• TECHNICAL TRANSLATION. English-language translations of foreign scientificand technical material pertinent to NASA’smission.

Specialized services that complement the STIProgram Office’s diverse offerings includecreating custom thesauri, building customizeddatabases, organizing and publishing researchresults . . . even providing videos.

For more information about the NASA STIProgram Office, see the following:

• Access the NASA STI Program Home Pageat http://www.sti.nasa.gov

• E-mail your question via the Internet tohelp@sti.nasa.gov

• Fax your question to the NASA AccessHelp Desk at 301–621–0134

• Telephone the NASA Access Help Desk at301–621–0390

• Write to: NASA Access Help Desk NASA Center for Information 7121 Standard Drive Hanover, MD 21076

Roy C. Tew and Rodger W. DysonGlenn Research Center, Cleveland, Ohio

Scott D. Wilson and Rikako DemkoSest, Inc., Middleburg Heights, Ohio

Overview 2004 of NASA-Stirling ConvertorCFD Model Development and RegeneratorR&D Efforts

NASA/TM—2004-213404

November 2004

National Aeronautics andSpace Administration

Glenn Research Center

Prepared for theSpace Technology and Applications International Forum (STAIF–2005)sponsored by the University of New Mexico’s Institute for Spaceand Nuclear Power Studies (UNM-ISNPS)Albuquerque, New Mexico, February 13–17, 2005

Acknowledgments

The work described in this paper was performed for NASA Headquarters, Science Mission Directorate (Code S)and Exploration Systems Mission Directorate (Code T).

Available from

NASA Center for Aerospace Information7121 Standard DriveHanover, MD 21076

National Technical Information Service5285 Port Royal RoadSpringfield, VA 22100

Trade names or manufacturers’ names are used in this report foridentification only. This usage does not constitute an officialendorsement, either expressed or implied, by the National

Aeronautics and Space Administration.

Available electronically at http://gltrs.grc.nasa.gov

NASA/TM—2004-213404 1

Overview 2004 of NASA-Stirling Convertor CFD Model Development and Regenerator R&D Efforts

Roy C. Tew* and Rodger W. Dyson

National Aeronautics and Space Administration Glenn Research Center Cleveland, Ohio 44135

Scott D. Wilson and Rikako Demko

Sest, Inc. Middleburg Heights, Ohio 44135

Abstract. This paper reports on accomplishments in 2004 in (1) development of Stirling-convertor CFD models at NASA Glenn and via a NASA grant, (2) a Stirling regenerator-research effort being conducted via a NASA grant (a follow-on effort to an earlier DOE contract), and (3) a regenerator-microfabrication contract for development of a “next-generation Stirling regenerator.” Cleveland State University is the lead organization for all three grant/contractual efforts, with the University of Minnesota and Gedeon Associates as subcontractors. Also, the Stirling Technology Company and Sunpower, Inc. are both involved in all three efforts, either as funded or unfunded participants. International Mezzo Technologies of Baton Rouge, Louisiana, is the regenerator fabricator for the regenerator-microfabrication contract. Results of the efforts in these three areas are summarized.

INTRODUCTION

A high-efficiency Stirling Radioisotope Generator (SRG) for use on potential NASA Space Science missions is being developed by the Department of Energy (DOE), Lockheed Martin, Stirling Technology Company (STC), and NASA Glenn Research Center (GRC). Potential missions include providing spacecraft onboard-electric power for deep-space missions or power for unmanned Mars rovers. GRC is also developing advanced technology for Stirling convertors, aimed at improving specific power and efficiency of the convertor and the overall power system. Performance and mass improvement goals have been established for second- and third-generation Stirling radioisotope power systems. Efforts are underway to achieve these goals, both in-house at GRC and via grants and contracts. These efforts include validation of a multi-dimensional (multi-D) Stirling computational-fluid-dynamics (CFD) model, high-temperature materials, advanced controllers, low-vibration techniques, advanced regenerators, and a lightweight convertor (Thieme, 2004; Schreiber, 2004). The objective of this paper is to report on the NASA multi-D code validation effort and NASA regenerator R&D efforts.

MULTI-DIMENSIONAL STIRLING MODEL DEVELOPMENT AND VALIDATION Mahkamov first reported on development of a multi-D model of a complete Stirling engine (Mahkamov, 2000). He recently reported on using the commercial Fluent code for Stirling engine 3-D modeling (Mahkamov, 2003). A NASA grant to Cleveland State University (CSU) for development of a multi-D Stirling engine model resulted in delivery of the 2-D CSUmod model to NASA Glenn (GRC) in 2003 (Ibrahim, 2004a). This model was developed using the CFD-ACE commercial code. The grant also included fabrication plus oscillating- and steady-flow visualization and testing of a 90-degree-turn test module at the University of Minnesota (UMN) (Adolfson, 2002, 2003). A new 180-degree-turn test module was designed and fabrication was begun at UMN before the grant ended on July 31, 2004. A new grant to CSU, and subcontractors UMN and Gedeon Associates, was awarded on Sept. 1, 2004 to continue multi-D code validation and support of multi-D code development.

* Phone: 216–433–8471, Fax: 216–433–6133, E-mail: Roy.C.Tew@nasa.gov

NASA/TM—2004-213404 2

In the last year, GRC substantially increased its level of in-house effort on CFD modeling of Stirling convertors. The 2-D model of STC’s ~55 We Technology Demonstration Convertor (TDC), developed based on the CSUmod model, was further developed (Wilson, 2004a). A Fluent commercial code license was purchased to complement the CFD-ACE commercial code. A 32 processor Microway computer cluster was purchased to enhance the capabilities provided by an earlier 8 processor Dell cluster (Dyson, 2004a; Wilson, 2004a). Dyson has also suggested new approaches for enhancing the speed and accuracy of future Stirling multi-D codes (Dyson, 2004a). The Stirling engine companies, STC and Sunpower, have both begun to make some use of multi-D Stirling models to support their engine-development efforts (Qiu, 2004; Wood, 2004a).

GRC Stirling-Convertor Model-Development Plans and Progress Dyson recently drafted a long-range Stirling analysis plan or “roadmap” for GRC’s Thermal Energy Conversion Branch (Dyson, 2004b). Some of the elements of this plan are outlined below. Also below, are summaries of plans and progress at GRC in the areas of Sage 1-D modeling and development of commercial multi-D transient models of Stirling convertors.

Stirling-Convertor Multi-D Model-Development Plans Dyson has suggested that GRC’s future multi-D CFD efforts should fall into the following five areas: (1) Help develop upgrades to current commercial transient codes (e.g., improve porous media models, improve conjugate heat transfer convergence). (2) Use available multi-D CFD codes for “engineering services” or engine component modeling, as needed; for example, the CFD-ACE code was used to model the cooling jacket of the laboratory version of STC’s TDC engine—since cooler-pressure-vessel-wall temperatures cannot easily be measured (Wilson, 2004b). (3) Develop a new transient code based on high-order accuracy numerical techniques to enhance speed of convergence and accuracy (Dyson, 2002, 2004a). (4) Consider development of 2-D and 3-D steady-periodic Stirling design codes, based on techniques similar to those used in Gedeon Associates’ Sage 1-D model (Gedeon, 1999). (4) Develop high-accuracy, fast-convergence 1-D, 2-D and 3-D transient thermodynamic models suitable for integration with GRC’s Stirling Dynamic Model or SDM (Regan, 2004; Lewandoski, 2004).

Dyson noted that most computational Stirling analysis techniques use low-order, finite-volume approaches because high-order accuracy is considered too difficult or expensive (Dyson 2004a). He also noted that new high-order-accuracy techniques overcome these earlier difficulties. To illustrate the benefits of high-order-accuracy techniques, Figure 1 shows wavelength error magnitude as a function of grid points per wave length (PPW) for numerical techniques of various orders of accuracy. The maximum wavenumber, for a geometric length, L, is the maximum number of waves (or the highest harmonic) that can be represented along the length, given a number of grid points, N. The plot shows that the PPW required to reduce the error magnitude to a given level, can be reduced by going to higher-order-accuracy techniques. Prof. Brad Egar of Arkansas State University, awarded an Arkansas State Grant funded by NASA, has proposed to work with GRC to investigate application of high-order-accuracy numerical techniques to Stirling modeling (Egar, 2004).

Sage 1-D Steady-Periodic Code Modeling Plans at GRC

GRC is devoting an increased level of effort to simulating several Stirling engines and a thermoacoustic engine with the Sage 1-D code. A GRC Sage Stirling-engine model exists for STC’s TDC engine. Sage models are to be set up for comparison with engine data from STC’s ~55 We TDCs 5&6, 7&8, 13&14 and 15&16 and from Sunpower’s

FIGURE 1. Wavelength Error Magnitude as a Function of Grid Points per Wavelength (PPW) for Numerical Techniques of Various Orders of Accuracy.

NASA/TM—2004-213404 3

~35 We “Buzz” engine (Wood, 2004b). It is also likely that a Sage model will be set up for Sunpower’s new Advanced Stirling Convertor (ASC) engine (Wood, 2004a). Also, the TRW/Northrup Grumman ~100 We Thermoacoustic Stirling Heat Engine will be modeled with Sage (Petach, 2004). Barry Penswick (Penswick 2004) proposed that GRC develop a new Sage model of the TDC hardware via a step-by-step approach to identify the impact on the Sage solution of adding increasing levels of complexity to the hardware models. The process involves keeping careful records of the overall cycle energy balance and the various losses as each new level of complexity is added to the Sage model. Careful tracking of the losses may identify areas where improvements can be made in the hardware. The process should be carried out for operating conditions corresponding to a well-documented engine test point; thus, model predictions at each stage of increasing complexity can be compared with engine performance test data. Penswick’s suggestion is currently being carried out at GRC (Demko 2004).

Progress and Problems in Development of a Commercial Code Multi-D Stirling Model at GRC

Wilson reported on progress at GRC in further development of a multi-D model of STC’s TDC (Wilson 2004a). He had added calculations of an overall energy balance and, with just a few engine cycles calculated, observed substantial improvement in the energy balance with each additional cycle. Plots of working space temperature, pressure, volume variations, and integrated component heat rates were shown over the first cycle to illustrate the transient nature of the solution. U-velocity vectors and contours from his report are reproduced in Figure 2.

This figure shows the current 2-D representation of the engine’s 3-D cooler and heater geometries. Thus the 3-D cooler and heater’s radial fins and flow passages are replaced by concentric-annular fins and flow passages of approximately equal flow area, hydraulic diameter and overall heat-transfer area. Then, having started with a 2-D axisymmetric representation of the engine, a requirement to use “arbitrary interfaces” to represent sliding-grid interfaces between the displacer and the appendix gap (not visible in Figure 2), forced use of a 3-D pie-slice representation of this equivalent 2-D geometry (since “arbitrary interfaces” were not available in CFD-ACE’s 2-D axisymmetric mode). The color velocity contours in Figure 2, plus the legend, indicate that the flow in the upper two cooler flow passages is in one direction, while the flow in the lower flow passage is in the opposite direction—with the displacer at it’s mean position (in traveling from the hot to the cold end). If this were the real engine geometry, if steady-periodic convergence had been achieved, and if this non-uniformity in flow persisted over a significant portion of the cycle—the results would indicate a substantial loss in performance. Note that the Sage 1-D code used to design the STC 55 We (TDC) engine assumes uniform axial flow and deviations from such uniformity imply a loss in performance.

Heater Regenerator

Cooler

Piston Face

Displacer Face

Displacer Face

FIGURE 2. NASA 2D TDC Model U-Velocity Vectors & Contours; Displacer at Mean Position

NASA/TM—2004-213404 4

Thus, to maximize the usefulness of a multi-D TDC engine model, it seems essential that the model be converted to the real 3-D engine geometry as soon as possible. Until now, however, the simpler faster-to-converge 2-D geometry has been used to address changes in the engine geometry from the CSUmod design (e.g., including an appendix-gap model, and solving grid-compression problems that resulted with the smaller TDC clearances at displacer top-dead-center and bottom-dead-center positions). Overall and component energy balances have also been incorporated. It is anticipated that transition to parallel computations on GRC’s Dell and Microway clusters will soon enhance the practicality of modeling the real engine (3-D) geometry. However, there are other problems to be addressed in use of commercial codes (CFD-ACE and Fluent in particular) to develop multi-D models of Stirling engines: (1) The equilibrium-porous-media models in the commercial codes must be replaced by non-equilibrium models, to allow regenerator solid and gas to be modeled with different temperatures. Several non-equilibrium models have been identified and tests at UMN (Niu 2004a, 2004b) are providing values of empirical coefficients (permeability, inertial coefficient, and thermal dispersion) necessary to implement these models. (2) A number of RANS (Reynolds Averaged Navier Stokes) turbulence models are available in the commercial codes. However, experimental tests in UMN test rigs have suggested that, at a given location in the engine, there is transition from laminar-to-turbulent flow and back over an engine cycle, and that these transitions occur at different times in different locations. In CFD-ACE, one must currently choose laminar flow everywhere in the overall domain being simulated, or a particular RANS model or LES (Large Eddy Simulation); in addition, Fluent has the capability for DES (Detached Eddy Simulation). RANS turbulence model calculations sometimes agree with laminar calculations when appropriate (Tew, 2003), however they cannot be reliably used to predict transitions between laminar and turbulent flow. CSU’s modeling of turbulence phenomena in the UMN test rigs, has shown that for one process, one RANS turbulence model (such as k-omega) produces the best agreement with the data. For another process, another RANS turbulence model (such as k-epsilon, or Chien’s model) produces the best agreement (Ibrahim, 2003a). Thus accurate modeling of laminar-turbulent transitional flow is a significant problem in multi-D modeling of Stirling engines. (3) The transient conjugate-heat-transfer problem (large-heat-capacity solids exchanging heat with small-heat-capacity gases), greatly extends the number of cycles and time required for convergence to a steady-periodic cycle. The problem was improved at CSU by solving for wall- and regenerator-solid-temperature distributions using a steady-state solution approach, with piston and displacer held stationary—then allowing the piston and displacer to move; but the cycles required for convergence is still large. If the requirement to converge the solid-wall-temperatures could be eliminated, then flow and temperature distributions in the gas might reach steady-periodic-cycle convergence in perhaps 4 to 6 cycles—based on modeling of MIT test rigs with assumed constant-inside-wall temperatures (Ebiana 2004). In an earlier transient 1-D code developed at GRC, the heat capacity of the pressure vessels walls was not modeled, but that of the regenerator matrix was (Tew, 1978, 1983). Convergence to periodic steady-state was a major problem until one of the authors, Jefferies, found a way to use the regenerator-solid heat transfer over one cycle to extrapolate convergence of the regenerator-solid temperature over the next cycle. Based on a brief summary, a recent cryocooler conference publication (Harvey, 2004), also appears to address a technique for speeding convergence of regenerator-matrix temperature. Some such approach might work in the multi-D codes. However, in CFD-ACE, the same enthalpy relaxation factors are currently applied to the entire solution domain, i.e. to solid and gas. The literature should be searched for approaches to accelerating temperature convergence of conjugate-heat-transfer solids. Such an approach would likely reduce the accuracy of the overall transient solution, but seems desirable in these early stages of trying to use these slow-converging complete-engine, time-marching, CFD models to reach a periodic-steady state.

NASA Grants for Multi-D Stirling Model Validation

A new grant to CSU, and subcontractors UMN and Gedeon Associates, was awarded on Sept. 1, 2004 to continue multi-D model validation and to support further model development. Details of these efforts follow.

Development of New Test Modules for Stirling-Convertor Model Validation

The new CSU multi-D code grant will emphasize validation of Stirling multi-D models. However, CSU will also continue to support GRC in trying to accelerate convergence of these models, and in trying to solve problem areas such as porous-media and laminar-turbulent-transition modeling.

NASA/TM—2004-213404 5

Discussions with engineers at STC and Sunpower led to CSU’s focusing on the flow between the heater and expansion space. UMN’s new cam-and-follower oscillating-flow generator for the 180-degree-turn test section is shown in Figure 3. A more detailed schematic of the test section is shown in Figure 4. Unlike the earlier 90-degree-turn test section (Adolfson, 2002, 2003), the new test section will accomplish a full 180 degree turning of the flow, as in real hardware, and will include heat transfer. The large size and low frequency (~1 Hz) will allow measurements with high spatial and temporal resolution. Dynamic similarity will be maintained by matching the dimensionless parameters of a pattern Stirling engine. Both flow visualization and detailed measurements of flow velocities and temperatures will be carried out. Heat transfer measurements will be made. Plans are for the heater to initially be a porous section, with later replacement by discrete flow passages and solid walls, so that 3-D flow and temperature fields can be quantified. Velocity and temperature measurements will be made with hot-wire anemometry and thermocouples, respectively. UMN has much experience in making these measurements in an oscillating-flow-regenerator test rig (Niu, 2002, 2003a, 2003b, 2003c). As in earlier UMN test sections, the new one exhausts into and extracts air from the room at 1 bar pressure and 298 K (about 25 C). When heat-transfer measurements are begun, the heat-addition surface temperature will be about 20 C above the flow temperature. Flow vortices and their behavior were an interesting feature of the earlier 90-degree-turn tests (Adolfson, 2002, 2003). It is anticipated that the new test section, which is a much better representation of real Stirling-engine hardware, will provide interesting insights into how expansion-space flow vortices can impact heat transfer in the expansion space and flow-uniformity in the heater. CSU will test a new Stirling-cooler test section (Figure 6) in the Stirling Laboratory Research Engine (SLRE), shown in Figure 5 (Hoehn, 1982). The alpha-configuration SLRE (two opposed pistons) has feature size and frequency similar to engines of interest (testing has been up to 20 Hz frequency). Temperature, pressure, and pressure-drop measurements will be made for CFD-test-rig model comparison. Hoshino recently used a two-thermocouple technique to make dynamic temperature measurements in the SLRE test facility (Hoshino, 2004). Since the phase angle between the two opposed pistons is adjustable from 0 to 180 degrees, the facility can be used to produce just oscillating-flow through the test section, or a combination of oscillating-flow and pressure level as when operated as a Stirling cooler. SLRE measurements in an earlier test section are reported in (Ibrahim, 2004d).

FIGURE 3. New Oscillating-Flow Generator and 180 Degree Test Section.

Displacer Heater

Head

Regenerator Shroud

FIGURE 4. Schematic of New 180 Degree Test Section.

Cylinder 1

Test Section Location

Cylinder 2

FIGURE 5. SLRE with “Alpha” Arrangement of 2 Cylinders/Pistons, Showing Test Section.

FIGURE 6. New SLRE Stirling Cooler Test Section.

NASA/TM—2004-213404 6

Multi-Dimensional Modeling of Stirling-Process Test Modules

As for earlier UMN test sections, CSU will develop CFD models of the new UMN 180-degree-turn test section, and the CSU SLRE-Stirling-cooler test section. Validation of these CFD models of Stirling-engine-like physical processes, should also help provide confidence in the CSUmod and TDC models of complete Stirling engines. Gedeon Associates’ Sage 1-D model will also be used to model the SLRE-Stirling-cooler test section. The Sage 1-D model, which is used in design of STC’s and Sunpower’s Stirling engines and coolers, provides useful benchmark solutions to assess the progress of multi-D Stirling models.

2nd Law Analysis for Stirling-Process Multi-Dimensional Models

A grant for developing a thermodynamic 2nd law analysis of 2-D CFD models of two Massachusetts Institute of Technology (MIT) test rigs ended Nov. 30, 2004 (Ebiana 2004). The two test rigs were a gas spring, and a modified version of the original gas-spring test rig with an annular heat exchanger mounted on top of the gas spring (Kornhauser, 1989). The work involved careful development of the 2-D models using the CFD-ACE code, including checking for independence of the periodic-steady-state solution to changes in spatial-grid sizes, time-step size, and number of cycles. Solution independence was checked via several approaches including superposition of variable profiles at the same time in sequential cycles, root-mean-square differences in variable values relative to a standard superimposed grid at the same time in sequential cycles, and tracking of the model energy balance from cycle to cycle. The CFD-ACE energy balances, and variable profile plots, were compared with predictions of the Sage code for the two test rigs. It is interesting to compare some of the CFD-ACE and Sage results: For the gas spring at 10 RPM, 192.4 kPa, and wall temperature of 294 K—Sage calculated a hysteresis loss (equal to power in and heat generated) of 0.434 W and CFD-ACE calculated a hysteresis loss of 0.470 W, or a difference of about 8 %. Whereas, for the gas spring+heat exchanger, Sage calculated a power in/net heat generated of 29.9 W and CFD-ACE calculated 53.2 W, or a difference of about 78 %. Note that Sage is a 1-D code, assumes uniform axial flow, and thus can’t directly account for effects of flow separation at changes in flow area. This uniform-flow assumption should be a much better assumption for the gas spring than for the gas spring+heat exchanger, where there is a major change in area from the “gas spring” cylinder to the much smaller flow area of the annular heat exchanger. Thus, one would suspect that agreement is much poorer between the two codes for the gas spring+heat exchanger because of the limitations of Sage in accounting for the effects of changes in area. However, Sage does include approximate equations accounting for pressure drop and heat transfer “end effects,” so it’s also possible that differences may also be due to CFD-ACE modeling errors (i.e. numerical, geometrical approximations, etc.). Both codes agree reasonably well with the experimental hysteresis loss data for the gas spring. Power in/net heat transfer data was not reported for the gas spring+heat exchanger test rig, unfortunately. Overall available-energy losses and entropy gains were calculated for the two test rigs, based on interactions of the two systems with the environment (Ebiana, 2004). The remainder of the grant will be devoted to making available energy and entropy calculations throughout the domain, so as to break down the overall available energy loss into those due to heat transfer, viscous friction, and mixing. The sum of these available energy losses should be equal to the overall available energy loss based on interactions of the systems with the environment—except for differences introduced by numerical error. So, differences in these two approaches to calculating overall available-energy loss may provide some indication of the overall numerical error inherent in the CFD calculations. Sage 1-D code performance and available energy losses will provide a benchmark calculation for comparison. A follow-on grant is anticipated which would apply the developed 2nd law analysis procedure to a complete multi-D Stirling engine model.

REGENERATOR RESEARCH AND DEVELOPMENT Two Stirling-convertor regenerator efforts are being funded by NASA: (1) One is a regenerator grant for regenerator experiments and CFD modeling for learning about various current-technology-regenerator losses, and attempting to identify approaches for their improvement; an important part of the effort has also been determination of empirical coefficients (permeability, inertial coefficient, thermal dispersion) needed for CFD modeling of regenerators.

NASA/TM—2004-213404 7

(2) The second effort is a NASA Research Award Contract for regenerator microfabrication; the goal of this effort is to fabricate a new “defined geometry” (non-random) regenerator, with improved performance, mostly by reducing pressure drop losses, and improved reliability relative to current-technology (wire screen, random fiber).

NASA Regenerator Research Grant and DOE Regenerator Research Contract

The two-year NASA regenerator research grant, begun on Oct. 1, 2003, is a follow-on effort to a previous three-year DOE regenerator research contract (Ibrahim, 2004b). The DOE effort involved measurements of temperatures and heat transfer within a large-scale matrix, and study of the effects of jetting from cooler tubes into the matrix; these measurements were made with a 90% porosity matrix, the same porosity as used in several small Stirling convertors of interest to NASA. Experimental plans for the NASA grant included direct measurements of thermal dispersion in the 90% porosity regenerator, a repeat of previous testing in a new 95% porosity regenerator, investigation of the effect of various heat exchanger tube exit geometries on jetting into the matrix, and heat transfer and pressure drop testing of random-fiber regenerators with porosities as high as ~95% in the NASA oscillating-flow test rig on loan to Sunpower. Direct measurements of thermal dispersion have been recently reported (Niu, 2004a). Measurements for determining permeability and inertial coefficient were previously reported, but revised measurements have recently been made (Niu, 2004b). The large-scale regenerator measurements have been made at UMN. CSU has been supporting the UMN measurements via CFD models of the UMN test rig, and using the data for model validation. This modeling has involved use of a macroscopic porous media model available in the CFD-ACE code. With appropriate choice of permeability and inertial coefficient, CSU has determined that the CFD-ACE porous media model does a relatively accurate job of modeling fluid flow in the large-scale UMN matrix (Ibrahim, 2003b). However, the CFD-ACE model assumes gas-solid thermal equilibrium. This is not adequate for Stirling-regenerator modeling. CSU identified one thermal-non-equilibrium porous-media model that they feel may be adequate for CFD modeling; they began to try to define some of the empirical coefficients in the model (i.e. to define a “closure” model) via computations with a microscopic model of a wire-screen representative-elementary-volume (Ibrahim, 2004c). Other forms of a thermal-non-equilibrium model were presented at the Porous Media Workshop hosted by GRC in Cleveland in April 2004 (Kaviany, 2004; Ayyaswamy, 2004). Also presented at the workshop was a non-equilibrium porous-media model used by David Gedeon in the Manifest 2-D Regenerator/Manifold Modeling Code (Gedeon, 1989).

NASA Research Award Regenerator Microfabrication Contract

The first year of the regenerator microfabrication contract ended August 31, 2004. Annual presentations for this and nine other competing NRA Award contracts have been made and decisions on continuation of each are pending. Initiation of the 2nd year of effort for successful awardees has been delayed, probably until passage of the FY2005 NASA budget. Successful 2nd year contractors may also receive awards for a 3rd year of effort. Encouraging progress was made during the 1st year of the regenerator microfabrication effort. Several approximations of a parallel-plate regenerator were chosen as potential candidates for a new microfabrication concept. Many potential manufacturers were surveyed (Sun, 2004). David Gedeon developed a revision of an existing regenerator figure of merit (by including thermal dispersion) and projected that the power and efficiency of a particular new Sunpower engine could be improved by 6-9% by using a new microfabricated regenerator (Gedeon, 2003, 2004); these comparisons were based on one design optimized for a random-fiber regenerator and designs optimized for a new microfabricated regenerator. Terry Simon and Yi Niu developed an argument supporting the validity of dynamic similitude in regenerators, to support testing of large-scale regenerators, as requested by NASA. After interviewing many potential manufacturers and concepts, requests for proposals were sent to only two manufacturers. Based on the submitted proposals, about 8 months into the first year, International Mezzo Technologies of Baton Rouge, LA was chosen to be the fabricator. The chosen manufacturing process, based on Mezzo’s expertise, is a combination of LIGA (lithography, electroplating and molding) and EDM (electric discharge machining). After Mezzo was chosen, a variation of an “involute” approximation of a parallel-plate regenerator was chosen for development. A solid model of the concept

NASA/TM—2004-213404 8

is shown in Figure 7. The current plan is to stack these disks with successive disks “flipped” in order to alternate the involute direction. This should help reduce axial conduction and allow for redistribution of flow between disks. Mezzo succeeded, in the short time available, to use LIGA to fabricate the EDM tool shown in Figure 8. The contractual effort also includes fabrication and testing of a Large-Scale-Mock-Up (LSMU) of the regenerator at UMN. There they will be able to test “flipped” and “unflipped” stacks of disks. In the third year of the effort, stacks of these disks are to be tested first in the Sunpower oscillating-flow test rig, and later in an actual Stirling engine. CONCLUDING REMARKS

The results of several advanced Stirling technology efforts have been summarized. The goals are improvements in Stirling convertor performance via (1) development and validation of multi-dimensional Stirling CFD models, (2) experimental and computational research to investigate regenerator fluid-flow and heat-transfer phenomena and define empirical coefficients for CFD modeling, and (3) development of a new improved regenerator via microfabrication. Progress and problems associated with continued development of two-dimensional Stirling engine models and the experiments being conducted to validate the models are reported. With Cleveland State University’s delivery of the CSUmod-Stirling-engine model to NASA and addition of more manpower to the NASA in-house Stirling analysis effort, NASA has increased its level of effort in Stirling computational analysis. However, it is anticipated that NASA will continue to need grant and contractual support in the areas of validation testing and further development of CFD techniques to improve the time efficiency and accuracy of multi-D Stirling modeling.

REFERENCES Adolfson, D. A., Simon, T. W., Ibrahim, M. B., and Gedeon, D., “Measurements in Oscillatory Flows with Separation in Support

of Stirling Engine Model Development, in proceedings of 37th Intersociety Energy Conversion Engineering Conference, Washington, D. C., IECEC 2002 Paper No. 20142, August 2002.

Adolfson, D.A., Simon, T.W., Ibrahim, M. B., and Gedeon, D., “Unsteady Fluid Dynamics Simulation of a Stirling Engine Heater Head,” in the proceedings of 1st International Energy Conversion Engineering Conference, Portsmouth, VA, Paper No. AIAA-2003-5934, August 2003.

Ayyaswamy, P.S., University of Pennsylvania, Private Communication, 2004 Demko, R., Sest Inc., Private Communication, 2004. Dyson, R. W., “Technique for Very-High-Order-Nonlinear Simulation and Validation,” J. Comp. Acoustics, Vol. 10, No. 2, pp.

211-229, 2002. Dyson, R. W., Wilson, S. D., and Tew, R. C., “Review of Computational Stirling Analysis Methods,” in proceedings of 2nd

International Energy Conversion Engineering Conference, Paper No. AIAA-2004-5582, Providence, RI, August 15-19, 2004a. Dyson, R. W., “Stirling Analysis Suggestions,” Private Communication , September 2004b. Ebiana, A., Savadekar, R., and Vallury, A., “2nd Law Analysis of Sage and CFD-ACE+ Models of MIT Gas Spring and “Two-

Space” Test Rigs,” in proceedings of 2nd International Energy Conversion Engineering Conference, Paper No. AIAA-2004-5778, Providence, RI, August 15-19, 2004.

Egar, B., Arkansas State University, Private Communication, 2004.

Solid Model

• 86 micron channels• 14 micron walls• 1 mm ring spacing• 0.5 mm thick

FIGURE 7. Chosen Involute Concept.

Stirling Engine Nickel Electrode

Beam35 microns wide200 microns long

FIGURE 8. Mezzo EDM Tool, fabricated via LIGA.

NASA/TM—2004-213404 9

Gedeon, D., “Manifest: A Computer Program for 2-D Flow Modeling in Stirling Machines,” NASA Contractor Report 182290, May 1989.

Gedeon, D., “Sage User’s Guide,” 3rd Edition, Gedeon Associates, 16922 South Canaan Road, Athens, OH 45701, 1999. Gedeon, D., Gedeon Associates, Private Communication, 2003. Gedeon, D., Gedeon Associates, Private Communication, 2004. Harvey, J. P., Kirkconnell, C. S. and Desai, P. V., “A Fast and Accurate Regenerator Numerical Model,” in proceedings of 13th

International Cryocooler Conference, New Orleans, LA, March 2004. Hoehn, F. W., “Stirling Laboratory Research Engine,” JPL Publication 82-13, 1982. Hoshino, T., Watanasirisuk, P., Ibrahim, M., and Simon, T., “Dynamic Temperature Measurements in Stirling Cycle Machines

using Two-Thermocouple Technique,” in proceedings of 2nd International Energy Conversion Engineering Conference, Paper No. AIAA-2004-5559, Providence, RI, August 15-19, 2004.

Ibrahim, M., Zhang, Z., Kembhavi, S., Simon, T., Tew, R., and Gedeon, D., “A Model of 90 Degree Turn Oscillatory Flow and Heat Transfer,” in proceedings of 1st International Energy Conversion Engineering Conference, Paper No. AIAA-2003-5934, August 17-21, 2003a.

Ibrahim, M. B., Rong, W., Simon, T. W., Tew, R., and Gedeon, D., “Microscopic Modeling of Unsteady Convective Heat Transfer in a Stirling Regenerator Matrix, “ in proceedings of 1st International Energy Conversion Engineering Conference, Portsmouth, VA, Paper No. AIAA-2003-6016, 2003b.

Ibrahim, M.B., Mittal, M., Simon, T., and Gedeon, D., “A 2-D CFD Model of a Free-Piston Stirling Engine for Space Applications with Annular Heat Exchangers,” in proceedings of 2nd International Energy Conversion Engineering Conference, Paper No. AIAA-2004-5583, Providence, RI, August 15-19, 2004a.

Ibrahim, M. B., Simon, T., Gedeon, D., and Tew, R., “Improving the Performance of the Stirling Convertor: Redesign of the Regenerator with Experiments, Computations, and Modern Fabrication Technology,” Final Report to U.S. Department of Energy under Award No. DE-FC36-00GO10627, March 2004b.

Ibrahim, M. B., Rong, W., Simon, T., and Gedeon, D., “Simulations of Flow and Heat Transfer inside Regenerators Made of Stacked Welded Screens using Periodic Cell Structures,” in proceedings of 2nd International Energy Conversion Engineering Conference, Paper No. AIAA-2004-5599, Providence, RI, August 15-19, 2004c.

Ibrahim, M. B., Wang, M., “Experimental Investigation of Oscillatory-Flow Pressure and Pressure Drop through Complex Geometries,” in proceedings of 2nd International Energy Conversion Engineering Conference, Paper No. AIAA-2004-5560, Providence, RI, August 15-19, 2004d.

Kaviany, M., University of Michigan, Private Communication, 2004. Kornhauser, A. A., “Gas-Wall Heat Transfer During Compression and Expansion,” Ph. D. Dissertation, Mech. Eng. Dept.,

Massachusetts Inst. of Technology, Cambridge, MA, 1989. Lewandowski, E.J. and Regan, T. F., “Overview of the GRC Stirling Convertor System Dynamic Model, “ in proceedings of 2nd

International Energy Conversion Engineering Conference, Paper No. AIAA-2004-5671, Providence, RI, August 15-19, 2004. Mahkamov, K. and Ingram, D. B., “Theoretical Investigations on the Stirling Engine Working Process,” in proceedings of 35th

Intersociety Energy Conversion Engineering Conference, AIAA, Las Vegas, NV, July 2000. Mahkamov, K. and Djumanov, D., “Three-Dimensional CFD Modeling of a Stirling Engine,” in proceedings of 11th

International Stirling Engine Conference, Rome, Italy, November 19-21, 2003. Niu, Y., Simon, T.W., Ibrahim, M.B., and Gedeon, D., “Oscillatory Flow and Thermal Field Measurements at the Interface

between a Heat Exchanger and a Regenerator of a Stirling Engine,” in proceedings of 37th Intersociety Energy Conversion Engineering Conference, Washington, D.C., IECEC 2002 Paper No. 20141, 2002.

Niu, Y., Simon, T.W., Ibrahim, M.B., and Gedeon, D., “Thermal Dispersion of Discrete Jets upon Entrance to a Stirling Engine Regenerator under Qscillatory Flow Conditions,” in proceedings of 6th ASME-JSME Thermal Engineering Joint Conference, Hapuna Beach, Hawaii Island, Hawaii, TED-AJ03-641, 2003a.

Niu, Y., Simon, T.W., Ibrahim, M.B., Tew, R., and Gedeon, D., “Measurements of Unsteady Convective Heat Transfer Rates within a Stirling Regenerator Matrix Subjected to Oscillatory Flow,” in proceedings of 1st International Energy Conversion Engineering Conference, Portsmouth, VA, Paper No. AIAA-2003-6013, 2003b.

Niu, Y., Simon, T.W., Ibrahim, M.B., Tew, R., and Gedeon, D., “JET Penetration into a Stirling Engine Regenerator Matrix with Various Regenerator-to-Cooler Spacings,” in proceedings of 1st International Energy Conversion Engineering Conference, Portsmouth, VA, Paper No. AIAA-2003-6014, 2003c.

Niu, Y., Simon, T., Gedeon, D. and Ibrahim, M., “On Experimental Evaluation of Eddy Transport and Thermal Dispersion in Stirling Regenerators,” in proceedings of 2nd International Energy Conversion Engineering Conference, Paper No. AIAA-2004-5646, Providence, RI, August 15-19, 2004a.

Niu, Y. and Simon, T., University of Michigan, Private Communication 2004b. Penswick, B., Sest Inc., Private Communication, 2004. Petach, M., Tward, E., and Backhaus, S., “Design and Testing of a Thermal to Electric Power Convertor Based on

Thermoacoustic Technology,” 2nd International Energy Conversion Engineering Conference, Paper No. AIAA-2004-5518, Providence, RI, August 15-19, 2004.

Qiu, S. and Augenblick, J. E., “Preliminary Computational Fluid Dynamics Modeling of STC Stirling Engines,” in proceedings of 2nd International Energy Conversion Engineering Conference, Paper No. AIAA-2004-5584, Providence, RI, August 15-19, 2004.

NASA/TM—2004-213404 10

Regan, T. F., and Lewandowski, E. J., “Application of the GRC Stirling Convertor System Dynamic Model,” in proceedings of 2nd International Energy Conversion Engineering Conference, Paper No. AIAA-2004-5672, Providence, RI, August 15-19, 2004.

Schreiber, J. G. and Thieme, L.G., “Accomplishments of the NASA GRC Stirling Technology Development Project,” in proceedings of 2ndInternational Energy Conversion Engineering Conference, AIAA-2004-5517, Providence, RI, August 15-19, 2004.

Sun, L., Mantell, S. C., Gedeon, D., and Simon, T., “A Survey of Microfabrication Techniques for Use in Stirling Engine Regenerators,” in proceedings of 2ndInternational Energy Conversion Engineering Conference, Paper No. AIAA-2003-5647, Providence, RI, August 15-19, 2004.

Tew, R., Jefferies, K., and Miao, D., “A Stirling Engine Computer Model for Performance Calculations,” DOE/NASA/1011-78/24, NASA TM-78884, July 1978.

Tew, R. C. Jr., “Computer Program for Stirling Engine Performance Calculations,” DOE/NASA/51040-42, NASA TM-82960, January, 1983.

Tew, R.C. Jr. and Ibrahim, M. B., “Two-Dimensional Compressible Non-Acoustic Modeling of Stirling Machine-Type Components,” AIAA Journal of Propulsion and Power, Vol. 19, No. 5, pp. 922-929, 2003.

Thieme, L. G. and Schreiber, J. G., “Advanced Technology Development for Stirling Convertors,” in proceedings of Space Technology and Applications International Forum (STAIF 2004), edited by M. El-Genk, AIP Conference Proceedings 654, Melville, New York, pp. 432-439, 2004.

Wilson, S.D., Dyson, R.W., Tew, R.C., and Ibrahim, M.B., “Multi-D CFD Modeling of a Free-Piston Stirling Convertor at NASA GRC,” in proceedings of 2nd International Energy Conversion Engineering Conference, Paper No. AIAA-2004-5673, Providence, RI, August 15-19, 2004a.

Wilson, S. D., “CFD Modeling of Energy Rejection Heat Exchanger for Stirling Technology Demonstration Convertor at NASA Glenn Research Center,” Preliminary Information Release (PIR) Internal Memorandum, NASA GRC-5490, August 2004b.

Wood, J. G. and Carroll, C., “Advanced Stirling Convertor Program Update,” in proceedings of 2nd International Energy Conversion Engineering Conference, Paper No. AIAA-2004-5522, Providence, RI, August 15-19, 2004a.

Wood, J. G. and Lane, N. W., “Progress Update on the Sunpower 35 We Stirling Convertor,” in proceedings of 2nd International Energy Conversion Engineering Conference, Paper No. AIAA-2004-5621, Providence, RI, August 15-19, 2004b.

This publication is available from the NASA Center for AeroSpace Information, 301–621–0390.

REPORT DOCUMENTATION PAGE

2. REPORT DATE

19. SECURITY CLASSIFICATION OF ABSTRACT

18. SECURITY CLASSIFICATION OF THIS PAGE

Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources,gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of thiscollection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 JeffersonDavis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503.

NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89)Prescribed by ANSI Std. Z39-18298-102

Form ApprovedOMB No. 0704-0188

12b. DISTRIBUTION CODE

8. PERFORMING ORGANIZATION REPORT NUMBER

5. FUNDING NUMBERS

3. REPORT TYPE AND DATES COVERED

4. TITLE AND SUBTITLE

6. AUTHOR(S)

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)

11. SUPPLEMENTARY NOTES

12a. DISTRIBUTION/AVAILABILITY STATEMENT

13. ABSTRACT (Maximum 200 words)

14. SUBJECT TERMS

17. SECURITY CLASSIFICATION OF REPORT

16. PRICE CODE

15. NUMBER OF PAGES

20. LIMITATION OF ABSTRACT

Unclassified Unclassified

Technical Memorandum

Unclassified

National Aeronautics and Space AdministrationJohn H. Glenn Research Center at Lewis FieldCleveland, Ohio 44135–3191

1. AGENCY USE ONLY (Leave blank)

10. SPONSORING/MONITORING AGENCY REPORT NUMBER

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES)

National Aeronautics and Space AdministrationWashington, DC 20546–0001

Available electronically at http://gltrs.grc.nasa.gov

November 2004

NASA TM—2004-213404

E–14912

WBS–22–972–20–01

16

Overview 2004 of NASA-Stirling Convertor CFD Model Developmentand Regenerator R&D Efforts

Roy C. Tew, Rodger W. Dyson, Scott D. Wilson, and Rikako Demko

Stirling engines; Computerized simulation; Unsteady flow; Convective heattransfer; Regenerators

Unclassified -UnlimitedSubject Category: 20 Distribution: Nonstandard

Prepared for the Space Technology and Applications International Forum (STAIF–2005) sponsored by the Universityof New Mexico’s Institute for Space and Nuclear Power Studies (UNM-ISNPS), Albuquerque, New Mexico,February 13–17, 2005. Roy C. Tew and Rodger W. Dyson, NASA Glenn Research Center; and Scott D. Wilson andRikako Demko, Sest, Inc., Middleburg Heights, Ohio 44135. Responsible person, Roy C. Tew, organization codeRPT, 216–433–8471.

This paper reports on accomplishments in 2004 in (1) development of Stirling-convertor CFD models at NASA Glennand via a NASA grant, (2) a Stirling regenerator-research effort being conducted via a NASA grant (a follow-on effort toan earlier DOE contract), and (3) a regenerator-microfabrication contract for development of a “next-generation Stirlingregenerator.” Cleveland State University is the lead organization for all three grant/contractual efforts, with theUniversity of Minnesota and Gedeon Associates as subcontractors. Also, the Stirling Technology Company andSunpower, Inc. are both involved in all three efforts, either as funded or unfunded participants. International MezzoTechnologies of Baton Rouge, Louisiana is the regenerator fabricator for the regenerator-microfabrication contract.Results of the efforts in these three areas are summarized.